Comparison of image fusion techniques in RGB & Kekre´s LUV color space

With the advent and augmentation of image processing, image fusion has emerged as a topic of great importance in many research areas such as automatic object detection, medical imaging, computer vision, remote sensing, robotics, military and law enforcement, image classification and so on. This dissertation discusses the basic concepts of image fusion, different spatial and frequency domain algorithms for image fusion. We propose techniques that make use of other simple spatial pixel methods in combination with different transforms & hybrid wavelets instead of only using only averaging method to fuse the images. The transform and wavelet based methods are implemented using row, column and full transforms. A comparative assessment of pixel and block level image fusion methods is performed using six image quality metrics. The objective is to find which method gives better results for multi-focus color image fusion in RGB and LUV color space.